Replacing text in a .nfo file

Question

I have a colors.nfo file where I want to replace variables and get a new generated nfo-file without losing the template.

There are ascii-signs which I don't know how to handle. Every time I load with file.open and replace the variables and write it to a new file, there are weird signs and the template is destroyed.

Here is an image of the file: https://i.imgur.com/8lqqXpg.png

Here is the uploaded file to handle with: Click to download -- Hope its okay. otherwise i will delete!

Hope you understand the problem. Want to replace the "%REPLACE1%", "%REPLACE2%" and "%REPLACE3%" with for example "BLACKGREY", "REDWHITE"....

• Tried to load it in a string with f.open
• after I replaced it with string.replace("%REPLACE1", "BLACKGREY")
• after I write a new file with f.write
• file is destroyed and the ascii signs are unreadable and the template is not like before

Code Example:

replaceString = []
f = open("colors.nfo")
for line in f:
    replaceString.append(line.rstrip())
f.close()
replaceColors = "\n".join(replaceString)
print(replaceColors.replace("%REPLACE1%", "BLACKGREY"))

Output:

ÛÛ³    [x] Yellow      [ ] Yellow       [ ] Yellow      ³ÛÛ
ÛÛ³    [x] Pink        [ ] Pink         [ ] %REPLACE3%  ³ÛÛ
ÛÛ³    [ ] Green       [ ] green        [ ] Green       ³ÛÛ
ÛÛ³    [ ] Red         [ ] red          [ ] Red         ³ÛÛ
ÛÛ³    [ ] Blue        [ ] blue         [ ] Blue        ³ÛÛ
ÛÛ³    [ ] Black       [ ] %REPLACE2%   [ ] black       ³ÛÛ
ÛÛ³    [ ] White       [ ] white        [ ] white       ³ÛÛ
ÛÛ³    [ ] grey        [ ] grey         [ ] grey        ³ÛÛ
ÛÛ³    [ ] brown       [ ] brown        [ ] brown       ³ÛÛ
ÛÛ³    [ ] BLACKGREY  [ ] orange       [ ] orange      ³ÛÛ
ÛÛ³    [ ] purple      [ ] purple       [ ] purple      ³ÛÛ

How it should be:

██│    [x] Yellow      [ ] Yellow       [ ] Yellow      │██
██│    [x] Pink        [ ] Pink         [ ] %REPLACE3%  │██
██│    [ ] Green       [ ] green        [ ] Green       │██
██│    [ ] Red         [ ] red          [ ] Red         │██
██│    [ ] Blue        [ ] blue         [ ] Blue        │██
██│    [ ] Black       [ ] %REPLACE2%   [ ] black       │██
██│    [ ] White       [ ] white        [ ] white       │██
██│    [ ] grey        [ ] grey         [ ] grey        │██
██│    [ ] brown       [ ] brown        [ ] brown       │██
██│    [ ] BLACKGREY   [ ] orange       [ ] orange      │██
██│    [ ] purple      [ ] purple       [ ] purple      │██

I don't want these "ÛÛ" in my new created file. I want to have the "blackboxes" like in the screen. Replacing is not the problem. Problem is the structure after loading the file into a string. when I write these string to a new file, the template does not look like in the screen shown.

Answer 1

There are two problems here

1. reading and writing the data correctly
2. preserving the structure

Reading and Writing

According to Wikipedia , .nfo files of this type are encoded with the cp437 text encoding. Therefore this encoding must be specified when reading and writing the file.

with open('colors.nfo', 'r', encoding='cp437') as f:
    ...

with open('colors.nfo', 'w', encoding='cp437') as f:
    ...

Alternatively, the file may be opened in binary mode, and all operations carried out using bytes rather than text.

with open('colors.nfo', 'rb') as f:
    ...

If no encoding is specified, Python uses the system default, probably cp1252 in this case.An 8-bit encoding like cp1252 is able to decode the file, but where cp437 decodes b'\xdb' as 'FULL BLOCK' (█), cp1252 decodes it as 'LATIN CAPITAL LETTER U WITH CIRCUMFLEX' (Û), corrupting the data as seen in the question.

Preserving Structure

The data has a fixed-width format, so care must be taken when replacing text that the target and replacement strings are both of equal length, otherwise the columns will not be aligned correctly.

target = '%REPLACE1%'
replacement = 'BLACKGREY'

delta = len(target) - len(replacement)

padding = ' ' * abs(delta)
if delta > 0:
    replacement += padding
else:
    target += padding

Solutions

Here's a complete script.

replacements = { 
        '%REPLACE1%': 'BLACKGREY',
        '%REPLACE2%': 'REDWHITE',
}   

with open('colors.nfo', encoding='cp437') as f:
    data = f.read()

for target, replacement in replacements.items():
    delta = len(target) - len(replacement)
    padding = ' ' * abs(delta)
    if delta > 0:
        replacement += padding
    else:
        target += padding

    data = data.replace(target, replacement)

with open('new-colors.nfo', 'w', encoding='cp437') as f:
    f.write(data)

Processing as binary data is the same, except that the files are opened and closed in binary mode, and strings are declared as bytes rather than str .

replacements = { 
        b'%REPLACE1%': b'BLACKGREY',
        b'%REPLACE2%': b'REDWHITE',
}   

with open('colors.nfo', 'rb') as f:
    data = f.read()

for target, replacement in replacements.items():
    delta = len(target) - len(replacement)
    padding = b' ' * abs(delta)
    if delta > 0:
        replacement += padding
    else:
        target += padding

    data = data.replace(target, replacement)

with open('new-colors.nfo', 'wb') as f:
    f.write(data)